Dynamic models, non-human essentials, free...
JD_Mortal
Posts: 760
Dynamic: Created from formula, following rules of a structure, and bonded with external components to create a new unique component that is "built to desire".
This is something I have always loved related to models and modeling. I do this, as well as others, manually. However, some things do not have to be manual.
Models, for the most part, are basic structures. They have rules to form and often have specific expectations. For a primitive example (no pun intended), I will use something we all use, daily. A dresser. (Ok, most of us use them daily.)
A dresser has a basic structure. The body which houses the drawers, provides structure for the top, sides, feet and trimmings, and the drawers which provide structure for its own trimmings, face and handles. The over-all structure, itself, confirms to standards that allow it to fit within doorways and provides some form of ergonomics. (Drawers not usually on the floor and deeper drawers being lower.)
For the most-part, this is the generic structure of many other things that are similar. Night-stands, trunks, cabinets, tool-chests, etc...
Though, as we are often inclined to be unique, it is common-practice to make every individual component as separate items, including the trillions of variations that are so subtle, you hardly notice them. Here is where my "suggestion" comes into play...
Just like with the human models, (having that basic structure and rules), other objects can severely gain from this form of "compression". Saving a morph, as opposed to a duplicate whole model, is a form of compression, using recycling of unchanged components. However, unlike the human models, the models I am talking about would be dynamic themselves. (Human models could be dynamic, but that is a whole cup of tea that I don't want to try to swallow at the moment.)
Here is the concept, using the dresser, again, as the example.
You select the dresser-form. (Assume it is that specific of an object.)
The dresser has ten inlay styles (3D shapes) available and ten graphic-styles.
The dresser has ten top-trim styles (3D contours), and ten graphic-styles similar to the inlay.
The dresser has five side-trim styles... (the four corners)... ditto to above...
The dresser has seven foot styles...
The dresser face is another component, bonded/inserted into this component...
- The face has options for 1,2,3 stacked rows of drawers
- The face has options for 1,2,3,4,5,6 columns of drawers, in each row
- The individual drawer-spaces have the options to have covers or be left as uncovered space
- There is an option for drawer-gaps
- There is an option for solid-cover drawers (no face and actually just skips this drawer in the column)
Back to the dresser... Now that it has a drawer sub-component to play with...
The dresser-drawers have ten cover-styles... (basic square, trimmed, inlay, rounded, decco...)
The dresser-drawers also have a sub-component style for handles (wood, brass, plastic, knob, latch)
The dresser-drawers also have a sub-component style-position for the handles (Center, left-n-right, three centered)
Optionally, some faces have specific variations. (Being swing-open with hinges, roll-out, being flip-down/up)
All of the above being one actual formulated object, created from individual components that simply obey structure rules. These also possibly being interactive for manipulation. (Pulling the dresser wide would not just scale it, but optionally create more drawers that auto-fit, based on the style selected. Raising the height would add more rows of drawers instead of just stretching the drawers out of proportion. Keeping all the trims of the same scale, if it was "stretched" instead of "expanded with more drawers".)
This can not be done with simple morphs, for most of the above... However, it is just a simple "adjustment" in the model to do all of this. The base-model itself (the components that are not "formulated"), consume little space.
In the end, you have one "object" that can be turned into a million possible unique objects/items. This also helps when it comes to making "matching sets" with "similar variations", as each sub-component is not actually bound to any one model, and can be used anywhere there is a "knob" or "foot" or "inlay-top" or "trim" or "drawer".
This is actually nothing new. CAD has used something similar to this for years, as-well-as google-sketchup-3D and many other programs. Often they are for stuff just as dull... Stairway generators, foundation generators, roofing generators, kitchen-cabinet generators...
This is also the reason I believe the "base foundations" should be free, with limited base-variations. This gives us the ability to create all the unique individual style components that others can use to "build" what they need, without us having to build a 42-piece matching set of furniture, or knives, or street-signs, or lamps, or... yadda-yadda-yadda...
There is even some cool "helper" options already available in daz, to aid in this. Same with hexagon and all other programs. (The script, nodes, and attaching meta-data that can be used to flag things like expanding segments and locked-scales, etc...)
This would, however, require an actual "registered component ID" and naming-convention, with a similar matched catalog for ease of locating and matching "compatible" stuff and "similar styles", and "versions". (Unlike how the catalog is now, where everything is sort-of jumbled into one giant resource. Though, there are flags to isolate things like "V4 compatible", which is OK, but honestly, anything that specific should be isolated into a catalog only for V4 stuff. This would require a whole area just for dynamic component bases and component parts. You don't want to search for feet, and get human feet and dresser feet in a general search.)
This could help give DAZ a little more of a tactical advantage over others, if it is done right. Not only for the users, but also for the content creators.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Room creator for furniture
Well, furniture would be one of the major ones that everyone gravitates to. Simply because many renderings are done inside, and in some form of living space. Hard to avoid life. xD
But this would work for anything...
Rings, Necklaces, Vehicles, Doors, Windows, Roads, Signs (trillions of them... Catalog just that and your HD is full.), Tables, Chairs, Lamps, Railings, Stairs, Walls (Brick, stone, standard, etc.), Trees, Wood-piles, Shoes, Weapons, Ships...
I would rather have three of each of the above, then have to fumble through 1000 of each of the above, to have enough variation to make my renderings "unique".
This would also include auto-detail, LOD selection, per object. So you are not rendering 10,000 polygon stop-signs all over town, except the one in front of you, at the dynamic bus-stop, on the side of your dynamic road, surrounded by your dynamic buildings. xD
If this was made exclusive to Hexagon and Daz, as "creation"... It helps to merge the two. The outputs would stay dynamic until "rendered" or exported. Once exported, it would just be another standard object to play with. Though, the version in DAZ and HEX, would just have to save the specific attributes, which are small number values that consume little space, like a cookie in a browser saves your site preferences.
Obviously not all objects could be this dynamic. Many unique items are just that... Unique. (Like a Victorian vampire furniture set, with more curves than a V6 model and specific non-repeating patterns within it, made for giants on an ice-planet.)
However, for an "essentials" pack of goodies... This would be a great addition and exposure for these new components. (Testing grounds)
This also lets people focus more on creating graphic-styles, while not having to fumble with the 3D object creation itself. Since that is a major hurdle that many of us have to struggle with. While reducing a lot of the "same object" clutter. (Standard primitive box, Square rug, Basic picture-frame, Stop sign... Must be like 800 of them, all exactly the same, wasted hard-drive space if you happen to own them all.)
Not to mention, it also increases the affordability of models for those of us on a tight budget. I could never bring myself to pay $5 for a 3D model of pliers, when I can get real ones for $1 at the store. Some of these prices are wonky. But that is a topic for another day. I have a simple solution for that too. xD (I have a solution for everything... Not that everyone will like the solutions. :P)
I think the biggest thing, besides the simplicity and ease of use, would be the speed gained. Not only speed in rendering, but speed in creating items, customizing and transferring styles where possible, and selecting the items needed. (Not having to dig through a 30,000 object file-system of nested folders, for sort-of-similar styles, priceless!)
After all, time is money and the more we save, the more we have to spend on others.
I would like something like that: props with morphs. Is there any vendor who frequently makes such things?
I do not know any, off the top of my head, but I am traversing the catalog slowly.
I may make a few component-based items like this, just without the integration. (Eg, you would have to manually position each item component to "construct" things.) Until I can dive into the scripting, to simplify it a little.
Any scripters who could help, would be appreciated, for helping to setup some demo-models. I am writing-up the "rules" for some components, at the moment. How they expand and "grow", the "positions" relative to the bases, and generic "deform rules" for the uniqueness. Not sure if the script has something like "random numbers", which would come-in handy for making random furniture.
Worst-case scenario, I will just write-up a simple external program that can inject the created objects into DAZ, or your favorite editor/studio.
http://www.daz3d.com/room-creator-version-2
Oh... Now I get your previous post... xD
Yea, that is similar to what I was talking about... (The way it functions as dynamic creation.)
I would have never found that on my own...
I'm pretty sure I'm not sure exactly what you're describing. :ohh:
But it did remind me of a program I've played with for years, and seems to offer some modelling abilities that resemble what you're describing. Or not.
http://www.groboto.com/v3/
If you search YouTube for "Groboto", you get LOTS of examples.
I still have version 1 of Groboto, it was free and create spectacular objects, but that eats all the resources of your PC.
I don't know the evolution of this program, you gives me desire for going to see…;-)
Its pretty much a niche program, but I had fun with it and actually created a few short animations that I liked. Its sort of like a procedural shader, i.e. a "procedural animation" where you set parameters rather than "posing" a figure. Not really up my "photo realistic" alley (Gen4 figures in Carrara), but a nice change of pace and *fairly* realistic if you fiddle with it. Meaning not just an ever-changing fractal or something like that. I don't recall any PC resource problem, but since animation is my one hobby that I spend money on, I buy pretty muscular machines - almost a necessity for the two day contests (48 Hour Film, etc.)
I never really tried to "model" anything, just used the presets to create something like an alien space station changing shape. But the demos and YouTube videos seem to show modelling is one use.
Completely off what I was saying... Just kidding...
No that is close too. Well, for part of it. The dynamic-shaping tool, but I wasn't going to discuss that in this post. (I was thinking of the legs, using a skewered lathe tool.)
The part I touched-on, was playing with existing forms. The part you touched on, which I actually left-out, was the component creation aspect. Those tools and that function would alter the components too much to make them functional as "structured models". (That is, with all mapped graphics and bumps, etc.)
The part I was talking about was slight altering of form, like stretching and scaling, but fixed along specific sections. Which allows you to scale-up the dresser without getting monster-sized dresser drawers and accents that blow it out of proportion, when compared to one that is slightly scaled-down, or unscaled. Then, being able to change the handles, trimmings, feet, tops, materials, and it changes them all to match, if desired.
However, a tool like that would be real handy for creating many of the individual components. Not sure how easy it is to use. Looks more scientific, with organic results, instead of being organic creation with organic results. (Like z-brush is organic creation and results.) However, for "furniture" ,something with a scientific and inorganic solid structure, a tool like that might be ideal.
Cool, we almost have a whole outline of a solid idea here...
Any other contributions?
Just playing with the many presets is pretty easy, and can give some interesting results. But as I mentioned, I have not gotten into any "modelling", so I can't say. I'm not a modeller anyway in any program, preferring to buy 3D elements and use them in short animations. So from that standpoint, "Groboto" has some appeal for me, i.e. using the presets to create short animations.